1 / 25

ATLAS Insertable B-Layer (IBL)

ATLAS Insertable B-Layer (IBL). LHCC Upgrade Review CERN, February 16, 2009 G. Darbo - INFN / Genova LHCC Review Agenda page: http://indico.cern.ch/conferenceDisplay.py?confId=52276. From Replacement to Insertable BL. One and half year ago (29/9/2007) the B-Layer Replacement Workshop.

ezhno
Télécharger la présentation

ATLAS Insertable B-Layer (IBL)

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. ATLAS Insertable B-Layer (IBL) LHCC Upgrade Review CERN, February 16, 2009 G. Darbo - INFN / Genova LHCC Review Agenda page: http://indico.cern.ch/conferenceDisplay.py?confId=52276

  2. From Replacement to Insertable BL • One and half year ago (29/9/2007) the B-Layer Replacement Workshop. • Serious difficulties found in the original idea of replacement: • Longer than foreseen shutdown of LHC for replacement (>1 year); • Activation of present detector ß safety issues; • Need to dismount service panels, disks, etc to access B-Layer ß time, risk of damage. • ATLAS appointed a B-Layer Task Force (BLTF): • Jan to Jun 2008 – BLTF convened every two weeks, chairs: A.Clark & G.Mornacchi; • July 2008 – BLTF Reported at Bern ATLAS Week, written document sent to the ATLAS EB. • An Insertable B-Layer (IBL)was the main recommendation of the BLTF: • ATLAS will appoint the IBL Project Leader this week; • Very motivated Pixel and Project Office groups, fully behind it.

  3. LHCC: Integrated Luminosity • Integrated luminosity affects detector life – Peak Luminosity affects R/O INTEGRATED LUMINOSITY Integrated Luminosity: 2008-17: 650 fb-1 2013-17: 550 fb-1 New injectors + IR upgrade phase 2 Before LHC accident. Need updated scenario for finalizing the IBL schedule Long Shutdown (8 months) Peak Luminosity: 2017: 3x1034 cm-2s-1 SLHC Start-up Early operation Collimation phase 2 Linac4 + IR upgrade phase 1 Ref: LHCC 1/7/2008 – Roland Garoby

  4. B-Layer Scenarios To maintain Pixel Detector performance with inserted layer, material budget is critical. Development of new local support structure with carbon-carbon foams. WH(120 Gev) SV1 εb=60% Light jets rejection SV1 εb=70% 2-layers R=3.5 cm and 8 cm 2-old layers ATLAS b-inserted as 4-layer R=3.5 cm b-replaced

  5. Main IBL Challenge • Beam pipe extraction and installation of the IBL + new Beam pipe • Complicated by material activation. • Fight everywhere for space: engineers are starting with making a real design. • Layout to fit into tight envelopes between present B-Layer and beam pipe. • Additional services (pipes, opto-fibers, electrical services). • Reduce beam pipe radius (Current internal r = 0.29). IBL assumes r = 0.25, with reduced isolation (from 8mm to 4mm). Smaller radius is investigated. • Keep low the IBL radiation-length: • Low radiation length (X0 = 60% of B-Layer) and smaller detector radius improve current Pixel detector physics performance (even with inefficient B-Layer). • Carbon-carbon foams with low density (ρ ~ 0.1÷0.2 g/cc) and reasonable thermal conductivity (K ~ 6÷18 W/m•K). • Head room in the cooling: low T, small fluid mass. • Electrical services low mass: Al (instead of Cu); high signal bandwidths. • Large active area in the modules (big FE chips). • Front-end chip and sensor design: • Higher radiation dose (200Mrad, 2x1016 neq/cm2), higher R/O bandwidths.

  6. Challenges: Beam Pipe Removal • Tools to dismount Beam Pipe support collars: • Remote access >3 m inside • Activated material – fast operation • Beam pipe must been supported from inside. • Tool has to compensate gravity bow (7m long pipe). Extraction tool Collar in remote position: 3m inside PP1 Ref: IBL Eng. Meeting 25/11/2008 – Yuri Gusakov

  7. B-Layer Replacement - Insertion Smaller radius B-layer to insert in the existing Pixel • 16-staves (current module “active” footprint gives hermetic coverage in phi, but current total width does not fit); IBL will be not shingled in Z (no space). Requires new smaller beam-pipe; beam-pipe Ø is the most important inputs • Pixel Modules: increase live area of the footprint: • New chip design (FE-I4) – live fraction, I/O bandwidth, 200 Mrad; • Sensor – increase radiation hard (smaller radius and ramping up LHC luminosity):3x1015 neq/cm2. New radiation dose simulation is going on with new release of Fluka. R&D and prototyping in 2009, construction 2010-2012;

  8. Possible Layouts Monostave With “agreed” FE-I4 size full coverage • Fighting for space to old b-layer against full coverage • 15 staves looks the most promising, at 35 or 36 mm radius, but mechanically is tight • We may not profit from smaller radius beampipe so much if we want full coverage (agreed module size). Tune FE-I4size? • We are also looking at several bistave options with inependent cooling circuits Flex Hybrid Curved (min R=1m) Straight • Castellated layer • 7 bistaves • 0 tilt • 32.5 IR, 41.5 OR • Bistave atractive: • Good clearace & overlap • More mechaniccally stable • Potenzial cooling redundance • However: • More difficult handling • Support may only be possible on one end 7mm offset Flex Hybrid Coverage Diagram Bistave Ref.: Neal Hartman

  9. Frontend Chip - FE-I4 FE-I3 @ R=5cm • Reasons for a new FE design: • Increase live fraction • New architecture to reduce inefficiencies (L=3xLHC) • New FE-I4 • Pixel size = 250 x 50 µm2 • Pixels = 80 x 336 • Technology = 0.13µm • Power = 0.5 W/cm2 • FE-I4 Design Status • Contribution from 5 laboratories. • Main blocks MPW submitted in Spring 2008. • Full FE-I4 Review: 2/3/3009 • Submission in Summer 3xLHC sLHC LHC inefficiency Hit prob. / DC 20.2mm ~200μm 7.6mm ~19 mm active IBM reticule 16.8mm 8mm active 2.8mm ~2mm Chartered reticule (24 x 32) FE-I3 74% FE-I4 ~89%

  10. Modules & Stave Arrangement • Two module options: • Single chip modules abut one against the next • Small sensor type: like 3D, active edge • Multi chip modules: chip look the same if using multi-chip modules • As present sensor size (~3xFE-I4) : like planar n-on-n • assuming no Z-shingling, no space. W-bond pads

  11. Sensors Options Planar (n-on-n) Pro’s n-on-n is a proven technology Lower Cdet -> lower noise, lower in-time threshold for same power settings in the FE. Partially depleted sensors collect charge Con’s No active edge (?) Guard ring -> dead area in Z, constraints the envelope (?) Charge collected at 600Vbias • Two “silicon” technologies considered: Planar and 3D sensors. • Could profit from 2 large “SLHC” R&D communities. 3D sensor • pro’s: • Larger charge collection after irradiation (but more power in the FE for same time-walk) • Active edge (butting modules) • Lower voltage (<150 V), power after irradiation • Con’s: • column Inefficiency at 90º • Higher Cdet • No experience in “scale” production • Several options and design flavours • Higher cost. Yield? • Other options? Diamonds could be a compatible technology • No cooling issues, low capacitance, no leakage current make them appealing… • Smaller community than silicon…

  12. External Services – ID Endplate • Final services arrangement necessitated improvisation • Cable over-length, mapping changes. Not all improvisations are documented ß in situ cross check is going on before closing of ATLAS. • Entering of nose region is critical even for few additional services, as available envelope is basically taken

  13. External Services • Installation of additional services for IBL is certainly not straight forward • Careful design on flange (and in ID endplate region) is necessary, which must combine • Verification in situ – happening now before closing ID end-plate & ATLAS • New design/drawings (in CATIA for flange) • Pipe routing for eventual CO2 cooling up to USA 15 should be o.k. • Radiation protection aspects have to be considered early enough All installation aspects of new IBL services have to be considered from the very beginning !

  14. Cooling – CO2 vs C3F8 • IBL cooling parameters: • 15 staves with 112W each ß Ptotal =1.68kW • Tsensor -25°C, ΔT to coolant ≤10°C ß Tcoolant -35°C • Options (limited by main constraint: develop time & working experience): • CO2: copy of the LHCb VELO system, similar in cooling power. • FC: present C3F8 system (after modifications). • Consider the new ATLAS and CERN reorganisation of the Cooling group: • ATLAS long term Upgrade and the improvement of present C3F8 system • Available Nikhef interest in contributing in the CO2 system (“cooling guru”).

  15. IBL Project Organization • IBL Project Leader (IBL PL) reporting to ATLAS and Pixel community: • Final round of IBL PL search – (probably) ATLAS CB nominates this week • Project Documents • Draft WBS exists: used to assign deliverables, costs and interest from ATLAS groups; • TDR in late 2009 (early 2010): TDR should not have options inside (sensor could be the exception); • Schedule: it will be agreed with CMS and LHC. Long shutdown for new triplet used to install. This will happen before the B-Layer will have seen life dose. • Funding Model • The overall model for the B-Layer Replacement was that this part of the detector was a “consumable”. • A dedicated line of funding, contained inside the Pixel M&O B, exists for the B-Layer Replacement. This will cover part of the costs. • First estimate would indicate a cost of about 7-8 MCHF, including new beampipe.

  16. Conclusions • Big progress in convergence to a B-Layer project: IBL • Feasibility studies on-going and “strawman” coming soon for several subparts • Organization structure will have soon Project Leader • Cost evaluation and funding model • Interest from groups to contribute. Open to Pixel and ATLAS • Challenging project • It will be an “assurance” for present B-Layer both for radiation end-of-life and for hard failures • Time scale is short, need optimization of design and prototyping but no time for basic R&D • Options should be kept small: decision between option is usually long and require parallel efforts.

  17. BACKUP SLIDES

  18. Critical Issues – Available Envelopes • Nominal Current B-Layer inner radius is just over 46 mm. • Envelope for B-Layer is 45.5 mm. • Assumed that is possible to reduce the beam pipe envelope • Reduce beam pipe isolation • Smaller beam pipe? R=25mm? • Need also clearance for beam pipe alignment (together with IBL) Current envelopes

  19. FE-I4 Architecture: Obvious Solution to Bottleneck • >99% or hits will not leave the chip (not triggered) • So don’t move them around inside the chip! (this will also save digital power!) • This requires local storage and processing in the pixel array • Possible with smaller feature size technology (130nm) Local Buffers pixels pixels Column pair bus Data transfer clocked at 20MHz bottleneck Sense amplifiers Buffer & serialiser Serial out trigger End of columnbuffer 64 deep trigger FE-I3 Column pair FE-I4 Column pair

  20. FE-I4_proto1 Collaboration • Participating institutes: Bonn, CPPM, Genova, LBNL, Nikhef. Bonn: D. Arutinov, M.Barbero, T. Hemperek, M. Karagounis. CPPM: D. Fougeron, M. Menouni. Genova: R. Beccherle, G. Darbo. LBNL: R. Ely, M. Garcia-Sciveres, D. Gnani, A. Mekkaoui. Nikhef: R. Kluit, J.D. Schipper FE-I4-P1 3mm LDORegulator 61x14 array Control Block ChargePump CapacitanceMeasurement SEU test IC 4mm DACs CurrentReference ShuLDO+trist LVDS/LDO/10b-DAC 4-LVDS Rx/Tx

  21. Internal Services – Stave Cable • Stave cable still on conceptual stage: • Cable using System Task force recommended signals + direct power. • Wire bonding MUST be done on stave • Many ideas (none developed to the end) • Single cable (conceptually like a circuit board to connect the FE chips) • Monolithic cable on top or on the bottom • Single cable also possible • Space is limited, what about reworking? End-of-stave connectors Unfolded cable Monolithic Cable on the Bottom

  22. From End-of-Stave (PP0) to PP1 “Strawman” Issuess: • X-section of LV cables • No active EOS -> 6m (FE-I4 to opto-board) • Interconnection space at PP0. • PP1 connectors… …more work to come to a design - EOS (PP0) Ref.: Marco Oriundo, Danilo Giugni, …

  23. LHCb-VELO Cooling System 2PACL (2-Phase Accumulator Controlled Loop) principle of cooling: • Liquid overflow => no mass flow control • Low vapor quality => good heat transfer • No local evaporator control, evaporator is passive in detector • Very stable evaporator temperature control with 2-phase accumulator (P4-5 = P7) Long distance P4-5 P7 5 Heat out Heat out Condenser 6 Heat in 4 Heat in 2 3 evaporator Restrictor Heat exchanger 1 2-Phase Accumulator Pump Vapor Liquid Pressure 2 3 2-phase 4 P7 5 1 6 Enthalpy Enthalpy

  24. R/O Links • Changes (strawman R/O from System Task force): • Down link (TTC) stay the same 40Mb/s (Manchester coding) – to 2 FE-I4 (?) – need clock multiplier in the FE-I4 (issues of SEU for clock multiplier). • Uplink use 160 Mb/s data+clock (8b/10b encode) – Single FE-I4 • Need new BOC design (substitute custom ASIC(s) with FPGA) • ROD could stay the same: reprogram FPGA (or new design for x2 links: need also 2 S-Links) • Use GRIN fibers (under rad-test for SLHC, or new Ericsson) • Opto-board at PP1 – need test of reliable electrical signal transmission (~4m) Present system

  25. Powering and DCS • Power scheme – Independent powering with PP1 regulators: • Use, on FE-I4 chip, x2 DC/DC and/or LDO conversion • DCS: similar implementation that today • Temperature sensors not on each module, smaller granularity • More than 600V if planar sensors? • Minor changes needed but it has a lot of different components to build/acquire. PP2 regulator 2.5V LDO 3.3V DC/DC FE-I4 Chip (the same we have now)‏ Remote sensing not needed for LDO (?) at PP1 for DC/DC

More Related